import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

link = 'https://raw.githubusercontent.com/sohaibbinmohsin/dm-project-boston-crime-analysis/main/Dataset.csv'
df = pd.read_csv(link, encoding = "ISO-8859-1", low_memory=False)

df.head(10)

	INCIDENT_NUMBER	OFFENSE_CODE	OFFENSE_CODE_GROUP	OFFENSE_DESCRIPTION	DISTRICT	REPORTING_AREA	SHOOTING	OCCURRED_ON_DATE	YEAR	MONTH	DAY_OF_WEEK	HOUR	STREET	Lat	Long	Location
0	I192068249	2647	Other	THREATS TO DO BODILY HARM	B2	280	NaN	8/28/15 10:20	2015	8	Friday	10	WASHINGTON ST	42.330119	-71.084251	(42.33011862, -71.08425106)
1	I192061894	1106	Confidence Games	FRAUD - CREDIT CARD / ATM FRAUD	C11	356	NaN	8/20/15 0:00	2015	8	Thursday	0	CHARLES ST	42.300605	-71.061268	(42.30060543, -71.06126785)
2	I192038828	1107	Fraud	FRAUD - IMPERSONATION	A1	172	NaN	11/2/15 12:24	2015	11	Monday	12	ALBANY ST	42.334288	-71.072395	(42.33428841, -71.07239518)
3	I192008877	1107	Fraud	FRAUD - IMPERSONATION	E18	525	NaN	7/31/15 10:00	2015	7	Friday	10	WINGATE RD	42.237009	-71.129566	(42.23700950, -71.12956606)
4	I182090828	1102	Fraud	FRAUD - FALSE PRETENSE / SCHEME	D4	159	NaN	12/1/15 12:00	2015	12	Tuesday	12	UPTON ST	42.342432	-71.072258	(42.34243222, -71.07225766)
5	I182090458	1107	Fraud	FRAUD - IMPERSONATION	C11	336	NaN	12/4/15 0:00	2015	12	Friday	0	NORTON ST	42.306265	-71.068646	(42.30626521, -71.06864556)
6	I182081063	1107	Fraud	FRAUD - IMPERSONATION	E18	478	NaN	12/1/15 12:25	2015	12	Tuesday	12	RUSKINDALE RD	42.268897	-71.108137	(42.26889666, -71.10813750)
7	I182074094	2629	Harassment	HARASSMENT	B2	258	NaN	9/14/15 9:31	2015	9	Monday	9	COLUMBIA RD	42.315142	-71.067047	(42.31514179, -71.06704709)
8	I182066132	2629	Harassment	HARASSMENT	B3	455	NaN	7/31/15 23:27	2015	7	Friday	23	RADCLIFFE ST	42.300202	-71.078354	(42.30020194, -71.07835353)
9	I182061268	3201	Property Lost	PROPERTY - LOST	NaN		NaN	6/15/15 0:00	2015	6	Monday	0	BERNARD	-1.000000	-1.000000	(-1.00000000, -1.00000000)

df.dtypes

INCIDENT_NUMBER         object
OFFENSE_CODE             int64
OFFENSE_CODE_GROUP      object
OFFENSE_DESCRIPTION     object
DISTRICT                object
REPORTING_AREA          object
SHOOTING                object
OCCURRED_ON_DATE        object
YEAR                     int64
MONTH                    int64
DAY_OF_WEEK             object
HOUR                     int64
STREET                  object
Lat                    float64
Long                   float64
Location                object
dtype: object

df.describe()

	OFFENSE_CODE	YEAR	MONTH	HOUR	Lat	Long
count	567623.000000	567623.000000	567623.000000	567623.000000	545093.000000	5.450930e+05
mean	2333.235131	2017.965038	6.624245	13.024004	36.494844	-6.129973e+01
std	1195.613580	1.897510	3.333956	6.333570	14.585577	2.448707e+01
min	100.000000	2015.000000	1.000000	0.000000	-1.000000	-7.117867e+01
25%	1102.000000	2016.000000	4.000000	9.000000	42.284070	-7.109272e+01
50%	3005.000000	2018.000000	7.000000	14.000000	42.316617	-7.107240e+01
75%	3201.000000	2020.000000	9.000000	18.000000	42.344513	-7.105576e+01
max	99999.000000	2022.000000	12.000000	23.000000	42.395042	5.250000e-08

Missing Values

num_rows = df.shape[0]
print('number of records: ',num_rows)
num_cols = df.shape[1]
print('number of attributes: ',num_cols)

number of records:  567623
number of attributes:  16

Number of missing values

print('Number of missing values:\n',df.isna().sum())
print('Total missing values: ',df.isna().sum().sum())

Number of missing values:
 INCIDENT_NUMBER             0
OFFENSE_CODE                0
OFFENSE_CODE_GROUP     214370
OFFENSE_DESCRIPTION         0
DISTRICT                 3421
REPORTING_AREA              0
SHOOTING               351798
OCCURRED_ON_DATE            0
YEAR                        0
MONTH                       0
DAY_OF_WEEK                 0
HOUR                        0
STREET                  11886
Lat                     22530
Long                    22530
Location                    0
dtype: int64
Total missing values:  626535

df['SHOOTING']= df['SHOOTING'].fillna(0)
df['REPORTING_AREA'].isna().sum()

0

df.dropna(subset=['Lat', 'Long', 'DISTRICT'], inplace=True)
df = df.bfill(axis ='columns')

df.shape

(541990, 16)

df.tail()

	INCIDENT_NUMBER	OFFENSE_CODE	OFFENSE_CODE_GROUP	OFFENSE_DESCRIPTION	DISTRICT	REPORTING_AREA	SHOOTING	OCCURRED_ON_DATE	YEAR	MONTH	DAY_OF_WEEK	HOUR	STREET	Lat	Long	Location
567618	222004519	617	LARCENY THEFT FROM BUILDING	LARCENY THEFT FROM BUILDING	A1		0	1/21/22 8:05	2022	1	Friday	8	ARLINGTON ST	0.0	0.0	(0, 0)
567619	222004517	520	BURGLARY - RESIDENTIAL	BURGLARY - RESIDENTIAL	D4		0	1/21/22 7:01	2022	1	Friday	7	WASHINGTON ST	0.0	0.0	(0, 0)
567620	222004516	3126	WARRANT ARREST - OUTSIDE OF BOSTON WARRANT	WARRANT ARREST - OUTSIDE OF BOSTON WARRANT	D4	167	0	1/21/22 7:25	2022	1	Friday	7	HARRISON AVE	0.0	0.0	(0, 0)
567621	122004001	3115	INVESTIGATE PERSON	INVESTIGATE PERSON	E13	912	0	1/19/22 9:30	2022	1	Wednesday	9	HEATH ST	0.0	0.0	(0, 0)
567622	149210376	3126	WARRANT ARREST - OUTSIDE OF BOSTON WARRANT	WARRANT ARREST - OUTSIDE OF BOSTON WARRANT	B2		0	1/2/22 0:00	2022	1	Sunday	0	WASHINGTON ST & ROXBURY ST\nBOSTON MA 02119\n...	42.3296	-71.08462	(42.32959998129787, -71.08462000476281)

print('test for NAN after filling the missing values: ',df.isnull().values.any() )

test for NAN after filling the missing values:  False

df.OFFENSE_CODE_GROUP.unique()

array(['Other', 'Confidence Games', 'Fraud', 'Harassment',
       'Criminal Harassment', 'Larceny', 'Property Lost',
       'Auto Theft Recovery', 'Larceny From Motor Vehicle',
       'Investigate Person', 'Police Service Incidents',
       'Warrant Arrests', 'Auto Theft', 'Restraining Order Violations',
       'Property Found', 'Robbery', 'Firearm Violations',
       'Medical Assistance', 'Simple Assault', 'Missing Person Located',
       'Missing Person Reported', 'Embezzlement',
       'Motor Vehicle Accident Response', 'Residential Burglary',
       'Counterfeiting', 'Prisoner Related Incidents', 'Vandalism',
       'Investigate Property', 'License Plate Related Incidents',
       'Aggravated Assault', 'Offenses Against Child / Family',
       'Firearm Discovery', 'Verbal Disputes', 'Search Warrants',
       'Violations', 'Landlord/Tenant Disputes', 'Commercial Burglary',
       'Evading Fare', 'Towed', 'Other Burglary', 'License Violation',
       'Drug Violation', 'Fire Related Reports', 'Disorderly Conduct',
       'Operating Under the Influence', 'Ballistics',
       'Property Related Damage', 'Assembly or Gathering Violations',
       'Recovered Stolen Property', 'Homicide', 'Liquor Violation',
       'Aircraft', 'Bomb Hoax', 'Service', 'Gambling', 'Prostitution',
       'Explosives', 'Arson', 'Phone Call Complaints',
       'Harbor Related Incidents', 'INVESTIGATE PERSON',
       'Burglary - No Property Taken', 'HOME INVASION', 'Manslaughter',
       'HUMAN TRAFFICKING', 'HUMAN TRAFFICKING - INVOLUNTARY SERVITUDE',
       'Biological Threat', 'Migrated Report - Other Part III',
       'Migrated Report - Auto Theft',
       'Migrated Report - Death Investigation',
       'Migrated Report - Kidnapping',
       'KIDNAPPING - ENTICING OR ATTEMPTED',
       'SICK/INJURED/MEDICAL - PERSON', 'VERBAL DISPUTE',
       'ASSAULT SIMPLE - BATTERY',
       'Migrated Report - Aggravated Assault/Aggravated Assault & Battery',
       'Migrated Report - Assault/Assault & Battery',
       'VAL - OPERATING WITHOUT LICENSE',
       'M/V - LEAVING SCENE - PROPERTY DAMAGE', 'WARRANT ARREST',
       'LARCENY THEFT OF MV PARTS & ACCESSORIES',
       'THREATS TO DO BODILY HARM',
       'MISSING PERSON - NOT REPORTED - LOCATED', 'TOWED MOTOR VEHICLE',
       'LARCENY ALL OTHERS', 'M/V ACCIDENT - PROPERTY Â\xa0DAMAGE',
       'M/V ACCIDENT - OTHER', 'DRUGS - OTHER', 'VANDALISM',
       'LARCENY THEFT OF BICYCLE', 'INVESTIGATE PROPERTY',
       'MISSING PERSON - LOCATED', 'HARASSMENT', 'PROPERTY - FOUND',
       'LARCENY THEFT FROM MV - NON-ACCESSORY', 'PROPERTY - LOST',
       'TRESPASSING', 'AUTO THEFT', 'VAL - OPERATING AFTER REV/SUSP.',
       'ANNOYING AND ACCOSTING', 'M/V ACCIDENT - OTHER CITY VEHICLE',
       'ASSAULT - AGGRAVATED', 'MISSING PERSON',
       'PROPERTY - ACCIDENTAL DAMAGE', 'FRAUD - FALSE PRETENSE / SCHEME',
       'LARCENY SHOPLIFTING', 'BURGLARY - RESIDENTIAL - FORCE',
       'FRAUD - CREDIT CARD / ATM FRAUD',
       'M/V ACCIDENT - PERSONAL INJURY',
       'SERVICE TO OTHER PD INSIDE OF MA.',
       'BURGLARY - RESIDENTIAL - NO FORCE',
       'VAL - OPERATING UNREG/UNINS Â\xa0CAR',
       'DRUGS - SICK ASSIST - HEROIN', 'Migrated Report - Fraud',
       'LARCENY PURSE SNATCH - NO FORCE ',
       'M/V ACCIDENT INVOLVING PEDESTRIAN - INJURY',
       'BURGLARY - RESIDENTIAL - ATTEMPT',
       'FIRE REPORT - HOUSE, BUILDING, ETC.',
       'ASSAULT - AGGRAVATED - BATTERY', 'DISORDERLY CONDUCT',
       'ROBBERY - OTHER', 'BURGLARY - OTHER - FORCE',
       'NOISY PARTY/RADIO-NO ARREST', 'SICK/INJURED/MEDICAL - POLICE',
       'LARCENY THEFT FROM BUILDING', 'SUDDEN DEATH', 'ROBBERY - STREET',
       'ASSAULT - SIMPLE', 'DRUGS - POSS CLASS A - HEROIN, ETC. ',
       'M/V ACCIDENT - INVOLVING BICYCLE - NO INJURY',
       'DRUGS - POSS CLASS A - INTENT TO MFR DIST DISP',
       'DRUGS - SALE / MANUFACTURING', 'LIQUOR - DRINKING IN PUBLIC',
       'INJURY BICYCLE NO M/V INVOLVED',
       'WEAPON - FIREARM - CARRYING / POSSESSING, ETC',
       'PROPERTY - STOLEN THEN RECOVERED', 'FORGERY / COUNTERFEITING',
       'BALLISTICS EVIDENCE/FOUND', 'FRAUD - IMPERSONATION',
       'M/V ACCIDENT - INVOLVING PEDESTRIAN - NO INJURY',
       'STOLEN PROPERTY - BUYING / RECEIVING / POSSESSING',
       'FUGITIVE FROM JUSTICE',
       'DRUGS - POSS CLASS C - INTENT TO MFR DIST DISP',
       'AUTO THEFT - LEASED/RENTED VEHICLE',
       'BURGLARY - COMMERICAL - FORCE', 'Car Jacking Firearm-Gas Station',
       'SEARCH WARRANT', 'DRUGS - POSS CLASS B - INTENT TO MFR DIST DISP',
       'LICENSE PREMISE VIOLATION', 'LANDLORD - TENANT SERVICE',
       'VAL - VIOLATION OF AUTO LAW - OTHER',
       'M/V - LEAVING SCENE - PERSONAL INJURY',
       'AUTO THEFT - MOTORCYCLE / SCOOTER',
       'M/V ACCIDENT - INVOLVING Â\xa0BICYCLE - INJURY',
       'M/V PLATES - LOST', 'PROPERTY - MISSING',
       'RECOVERED - MV RECOVERED IN BOSTON (STOLEN OUTSIDE BOSTON)',
       'M/V ACCIDENT - POLICE VEHICLE',
       'INVESTIGATION FOR ANOTHER AGENCY',
       'OPERATING UNDER THE INFLUENCE ALCOHOL', 'FRAUD - WELFARE',
       'FIREARM/WEAPON - FOUND OR CONFISCATED', 'EVADING FARE',
       'DEATH INVESTIGATION', 'DRUGS - POSS CLASS B - COCAINE, ETC.',
       'VAL - OPERATING W/O AUTHORIZATION LAWFUL',
       'DRUGS - POSS CLASS D - INTENT TO MFR DIST DISP',
       'DRUGS - SICK ASSIST - OTHER NARCOTIC',
       'VIOLATION - CITY ORDINANCE', 'ROBBERY - COMMERCIAL',
       'BURGLARY - COMMERICAL - NO FORCE',
       'ANIMAL CONTROL - DOG BITES - ETC.', 'GRAFFITI',
       'WEAPON - FIREARM - OTHER VIOLATION', 'EXTORTION OR BLACKMAIL',
       'REPORT AFFECTING OTHER DEPTS.', 'FIRE REPORT - CAR, BRUSH, ETC.',
       'Car Jacking Firearm-Bank', 'FRAUD - WIRE',
       'DRUGS - CLASS B TRAFFICKING OVER 18 GRAMS',
       'BURGLARY - OTHER - ATTEMPT', 'WEAPON - OTHER - OTHER VIOLATION',
       'EMBEZZLEMENT', 'WEAPON - OTHER - CARRYING / POSSESSING, ETC',
       'DRUGS - CLASS A TRAFFICKING OVER 18 GRAMS',
       'INTIMIDATING WITNESS', 'Car Jacking Firearm-Miscellaneous',
       'HARBOR INCIDENT / VIOLATION', 'ANIMAL INCIDENTS',
       'DEMONSTRATIONS/RIOT', 'DRUGS - CLASS D TRAFFICKING OVER 50 GRAMS',
       'BURGLARY - COMMERICAL - ATTEMPT', 'ROBBERY - HOME INVASION',
       'SUICIDE / SUICIDE ATTEMPT', 'OTHER OFFENSE',
       'DRUGS - POSS CLASS C', 'DRUGS - POSS CLASS D',
       'DISTURBING THE PEACE', 'DRUGS - POSSESSION',
       'DRUGS - SICK ASSIST - OTHER HARMFUL DRUG',
       'PROSTITUTION - SOLICITING', 'BURGLARY - OTHER - NO FORCE',
       'LARCENY PICK-POCKET', 'VIOLATION - HAWKER AND PEDDLER',
       'MURDER, NON-NEGLIGIENT MANSLAUGHTER', 'TRUANCY / RUNAWAY',
       'BIOLOGICAL THREATS', 'EXPLOSIVES - POSSESSION OR USE',
       'PROPERTY - LOST THEN LOCATED',
       'OPERATING UNDER THE INFLUENCE DRUGS', 'DRUGS - POSS CLASS E',
       'ARSON', 'AIRCRAFT INCIDENTS',
       'SERVICE TO OTHER PD OUTSIDE OF MA.',
       'Car Jacking Firearm-Highway', 'POSSESSION OF BURGLARIOUS TOOLS',
       'DRUGS - POSSESSION OF DRUG PARAPHANALIA', 'ROBBERY - CAR JACKING',
       'ANIMAL ABUSE', 'FIRE REPORT/ALARM - FALSE',
       'LIQUOR LAW VIOLATION', 'AFFRAY',
       'DANGEROUS OR HAZARDOUS CONDITION', 'ROBBERY - BANK',
       'DRUGS - POSS CLASS E - INTENT TO MFR DIST DISP',
       'EXPLOSIVES - TURNED IN OR FOUND', 'CHINS',
       'PROTECTIVE CUSTODY / SAFEKEEPING', 'PROPERTY - CONCEALING LEASED',
       'CRIMINAL HARASSMENT',
       'DRUGS - CONSP TO VIOL CONTROLLED SUBSTANCE', 'BOMB THREAT',
       'Larceny-Theft From Buildings', 'RECOVERED STOLEN PLATE',
       'OBSCENE PHONE CALLS', 'WEAPON - FIREARM - SALE / TRAFFICKING',
       'KIDNAPPING/CUSTODIAL KIDNAPPING',
       'PRISONER - SUICIDE / SUICIDE ATTEMPT', 'FIREARM/WEAPON - LOST',
       'NOISY PARTY/RADIO-ARREST', 'MANSLAUGHTER - VEHICLE - NEGLIGENCE',
       'PROSTITUTION', 'FIREARM/WEAPON - ACCIDENTAL INJURY / DEATH',
       'LARCENY THEFT FROM COIN-OP MACHINE', 'CONSPIRACY EXCEPT DRUG LAW',
       'HARASSMENT/ CRIMINAL HARASSMENT', 'PROPERTY - LOST/ MISSING',
       'SERVICE TO OTHER AGENCY', 'BURGLARY - COMMERICAL',
       'LANDLORD - TENANT', 'M/V ACCIDENT - PROPERTY DAMAGE',
       'M/V ACCIDENT - INVOLVING PEDESTRIAN - INJURY', 'ROBBERY',
       'VAL - VIOLATION OF AUTO LAW', 'BURGLARY - RESIDENTIAL',
       'M/V ACCIDENT - INVOLVING BICYCLE - INJURY',
       'RECOVERED - MV RECOVERED IN BOSTON (STOLEN IN BOSTON) MUST BE SUPPLEMENTAL',
       'BREAKING AND ENTERING (B&E) MOTOR VEHICLE',
       'Migrated Report - Motor Vehicle Crash',
       'Migrated Report - Vandalism/Destruction of Property',
       'Migrated Report - Other Larceny',
       'Migrated Report - Investigate Person',
       'Migrated Report - Other Part II',
       'Migrated Report - Affray/Disturbing the Peace/Disorderly Conduct',
       'Migrated Report - Burglary/Breaking and Entering',
       'Migrated Report - Larceny From MV',
       'Migrated Report - Drugs - Possession/Manufacturing/Distribute',
       'Migrated Report - Investigate Property',
       'Migrated Report - Auto Law Violation',
       'Migrated Report - Robbery',
       'Migrated Report - Counterfeiting/Forgery',
       'Migrated Report - Weapons Violation',
       'Migrated Report - Injured/Medical/Sick Assist',
       'Migrated Report - Criminal Homicide',
       'Migrated Report - Stolen Property',
       'Migrated Report - Other Auto Theft',
       'Migrated Report - Embezzlement', 'FIRE REPORT', 'SICK ASSIST',
       'VIOLATION - RESTRAINING ORDER (NO ARREST)',
       'DRUGS - POSSESSION/ SALE/ MANUFACTURING/ USE',
       'WARRANT ARREST - OUTSIDE OF BOSTON WARRANT',
       'LIQUOR/ALCOHOL - DRINKING IN PUBLIC',
       'SICK ASSIST - DRUG RELATED ILLNESS',
       'WEAPON VIOLATION - CARRY/ POSSESSING/ SALE/ TRAFFICKING/ OTHER',
       'ANIMAL INCIDENTS (DOG BITES, LOST DOG, ETC)',
       'LARCENY PURSE SNATCH - NO FORCE',
       'WARRANT ARREST - BOSTON WARRANT (MUST BE SUPPLEMENTAL)',
       'OPERATING UNDER THE INFLUENCE (OUI) ALCOHOL',
       'BREAKING AND ENTERING (B&E) MOTOR VEHICLE (NO PROPERTY STOLEN)',
       'VIOLATION - HARASSMENT PREVENTION ORDER',
       'OPERATING UNDER THE INFLUENCE (OUI) DRUGS',
       'DISTURBING THE PEACE/ DISORDERLY CONDUCT/ GATHERING CAUSING ANNOYANCE/ NOISY PAR',
       'CHILD REQUIRING ASSISTANCE (FOMERLY CHINS)',
       'VIOL. OF RESTRAINING ORDER W ARREST', 'DRUNKENNESS',
       'Evidence Tracker Incidents',
       'PROSTITUTION - ASSISTING OR PROMOTING',
       'KIDNAPPING/CUSTODIAL KIDNAPPING/ ABDUCTION',
       'Justifiable Homicide', 'MANSLAUGHTER - NEGLIGENCE'], dtype=object)

import folium
from folium import Marker
m_2 = folium.Map(location=[42.32,-71.0589], tiles='cartodbpositron', zoom_start=12)

daytime_robberies = df[((df.OFFENSE_CODE_GROUP == 'Robbery') & (df.HOUR.isin(range(9,18))))]
# Add points to the map
for idx, row in daytime_robberies.iterrows():
    Marker([row['Lat'], row['Long']], popup=row['DISTRICT']).add_to(m_2)

# Display the map
m_2

Make this Notebook Trusted to load map: File -> Trust Notebook

import folium
from folium import Marker
m_3 = folium.Map(location=[42.32,-71.0589], tiles='cartodbpositron', zoom_start=12)

drunk = df[(df.OFFENSE_CODE_GROUP == 'DRUNKENNESS')]
# Add points to the map
for idx, row in drunk.iterrows():
    Marker([row['Lat'], row['Long']], popup=row['DISTRICT']).add_to(m_3)

# Display the map
m_3

Make this Notebook Trusted to load map: File -> Trust Notebook

m_4 = folium.Map(location=[42.32,-71.0589], tiles='cartodbpositron', zoom_start=12)

Harr = df[(df.OFFENSE_CODE_GROUP == 'Harassment')]
# Add points to the map
for idx, row in Harr.iterrows():
    Marker([row['Lat'], row['Long']]).add_to(m_4)

# Display the map
m_4

Make this Notebook Trusted to load map: File -> Trust Notebook

Changing types of attributes

from sklearn.preprocessing import LabelEncoder
gle = LabelEncoder()
columns = ['INCIDENT_NUMBER', 'OFFENSE_CODE_GROUP', 'DISTRICT', 'SHOOTING', 'OCCURRED_ON_DATE', 'DAY_OF_WEEK', 'Location', 'STREET']
for c,col in enumerate(columns):
    labels = gle.fit_transform(df[col].astype(str))
    df[col] = labels
df.head()

	INCIDENT_NUMBER	OFFENSE_CODE	OFFENSE_CODE_GROUP	OFFENSE_DESCRIPTION	DISTRICT	REPORTING_AREA	SHOOTING	OCCURRED_ON_DATE	YEAR	MONTH	DAY_OF_WEEK	HOUR	STREET	Lat	Long	Location
0	437999	2647	133	THREATS TO DO BODILY HARM	6	280	0	287775	2015	8	0	10	2610	42.330119	-71.084251	16151
1	432189	1106	247	FRAUD - CREDIT CARD / ATM FRAUD	8	356	0	275894	2015	8	4	0	5975	42.300605	-71.061268	4377
2	410464	1107	14	FRAUD - IMPERSONATION	0	172	0	392471	2015	11	1	12	3691	42.334288	-71.072395	18511
3	382241	1107	14	FRAUD - IMPERSONATION	2	525	0	248106	2015	7	0	10	3362	42.237009	-71.129566	2447
4	361918	1102	14	FRAUD - FALSE PRETENSE / SCHEME	11	159	0	179	2015	12	5	12	2062	42.342432	-71.072258	22398

Correlation

Pearson Method

pcorr = df.corr(method='pearson')
print(pcorr)

                    INCIDENT_NUMBER  OFFENSE_CODE_GROUP  DISTRICT  SHOOTING  \
INCIDENT_NUMBER            1.000000           -0.035080 -0.000502  0.005544   
OFFENSE_CODE_GROUP        -0.035080            1.000000  0.000846 -0.017243   
DISTRICT                  -0.000502            0.000846  1.000000  0.007623   
SHOOTING                   0.005544           -0.017243  0.007623  1.000000   
OCCURRED_ON_DATE          -0.071468           -0.005393 -0.002008  0.002724   
DAY_OF_WEEK                0.001444            0.003476  0.003925 -0.002974   
STREET                     0.004485           -0.000509  0.035064 -0.003109   
Location                  -0.204209            0.000389 -0.202385 -0.020227   

                    OCCURRED_ON_DATE  DAY_OF_WEEK    STREET  Location  
INCIDENT_NUMBER            -0.071468     0.001444  0.004485 -0.204209  
OFFENSE_CODE_GROUP         -0.005393     0.003476 -0.000509  0.000389  
DISTRICT                   -0.002008     0.003925  0.035064 -0.202385  
SHOOTING                    0.002724    -0.002974 -0.003109 -0.020227  
OCCURRED_ON_DATE            1.000000     0.002525 -0.000428  0.039201  
DAY_OF_WEEK                 0.002525     1.000000 -0.001278 -0.003158  
STREET                     -0.000428    -0.001278  1.000000  0.023391  
Location                    0.039201    -0.003158  0.023391  1.000000  

plt.matshow(pcorr)
plt.show()

Kendall Method

kcorr = df.corr(method='kendall')
print(kcorr)

                    INCIDENT_NUMBER  OFFENSE_CODE_GROUP  DISTRICT  SHOOTING  \
INCIDENT_NUMBER            1.000000           -0.028901  0.000345  0.011805   
OFFENSE_CODE_GROUP        -0.028901            1.000000 -0.002290 -0.016327   
DISTRICT                   0.000345           -0.002290  1.000000 -0.002069   
SHOOTING                   0.011805           -0.016327 -0.002069  1.000000   
OCCURRED_ON_DATE          -0.013665           -0.002947 -0.000490  0.001863   
DAY_OF_WEEK                0.001053            0.001727  0.002270 -0.002449   
STREET                     0.002955            0.000127  0.024361 -0.000807   
Location                  -0.152945            0.004905 -0.098287 -0.021427   

                    OCCURRED_ON_DATE  DAY_OF_WEEK    STREET  Location  
INCIDENT_NUMBER            -0.013665     0.001053  0.002955 -0.152945  
OFFENSE_CODE_GROUP         -0.002947     0.001727  0.000127  0.004905  
DISTRICT                   -0.000490     0.002270  0.024361 -0.098287  
SHOOTING                    0.001863    -0.002449 -0.000807 -0.021427  
OCCURRED_ON_DATE            1.000000     0.001931 -0.000311  0.028930  
DAY_OF_WEEK                 0.001931     1.000000 -0.000918 -0.002309  
STREET                     -0.000311    -0.000918  1.000000  0.013075  
Location                    0.028930    -0.002309  0.013075  1.000000  

plt.matshow(kcorr)
plt.show()

Most Frequently Occuring Incident

print(df['OFFENSE_CODE'].value_counts())

3115    36406
3006    29493
3831    27947
1402    25470
3114    21427
        ...  
315         1
112         1
1105        1
2500        1
122         1
Name: OFFENSE_CODE, Length: 252, dtype: int64

From above, we can see that incident with offense code 3115 occured most. So, now we will find out its description.

ind = df.index[df['OFFENSE_CODE'] == 3115].tolist()
offense_desc = list()

for i in ind:
  offense_desc.append(df['OFFENSE_DESCRIPTION'][i])
print('Most occuring incidents is: ', set(offense_desc))

Most occuring incidents is:  {'INVESTIGATE PERSON'}

District Wise Analysis

district_incident_table = (df['DISTRICT'].value_counts()).to_frame()
district_incident_table.columns = ['Incidents']
district_incident_table.index.name = 'District'
district_incident_table

	Incidents
District
B2	87461
C11	75149
D4	73402
B3	63335
A1	62367
C6	42396
D14	36743
E18	31529
E13	31350
E5	24673
A7	23643
A15	11762
External	392

fig = plt.figure()
ax = fig.add_axes([0,0,1,1])
ax.bar(district_incident_table.index,district_incident_table['Incidents'])
plt.show()

#sns.boxplot(df['YEAR'])
sns.boxplot(df['Location'])

/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variable as a keyword arg: x. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.
  FutureWarning

<matplotlib.axes._subplots.AxesSubplot at 0x7f82dd384350>